up vote 19 down vote favorite
3

Possible Duplicate:
Write an Elevator Pitch / Tagline

Note:

We are closing this domain naming thread. It is asking the entirely wrong question. See this blog post for details: Domain Names: Wrong Question

We're going to keep the name stats.stackexchange.com. But we WILL be setting up redirects from the more "popular" domains names (e.g. seasonedadvice.com to cooking.stackexchange.com, basicallymoney.com to money.stackexchange.com, and others as we go through the list).

New question: "Write and Elevator Pitch / Tagline!"

Click here to contribute ideas and vote.

[original message text below]

Continuing with my previous question taken directly from the 7 Essential Meta Questions of Every Beta, here's another. What should we call our site (and linked to that what should our domain name be)?

Blog post about choosing domain.

First round ended; the second will be announced soon.

Here is a discussion about how the second round should look.

There are still new proposals comming! To see them click "newest" tab below: alt text

link
3  
Just to emphasize one of my overarching views on the name: the site should be about "data analysis", and that should include all aspects of statistics (yes, including both bayesian and frequentist), visualization, machine learning/data mining, "data science", etc. The title should include all these camps. – Shane Jul 26 '10 at 14:29
feedback

closed as exact duplicate by Robert Cartaino♦ Oct 9 '10 at 2:57

This question covers exactly the same ground as earlier questions on this topic; its answers may be merged with another identical question. See the FAQ for guidance on how to improve it.

49 Answers

1 2
up vote 12 down vote

ConfidenceRegion.com

(domain name available)

link
6  
Not bad -- a place for giving people confidence by answering their questions, as well as the obvious statistical meaning. – Rob Hyndman Aug 12 '10 at 11:31
3  
Dig it. It's not banal like StatsOverflow and should have meaning for the broad community of data analysts (statistics, machine learning, scientists, etc). Come on folks, vote it up! – ars Aug 19 '10 at 17:37
show 2 more comments
feedback
up vote 11 down vote

CrossValidated.com

link
2  
@mbq: I like your variation (made the change). – Shane Jul 24 '10 at 10:20
5  
I like this one the most at this point, because it not only covers a broadly applied technique, but also in some sense describes what we do on the site (i.e. have many different views on the same question to find the best solution). – Shane Jul 26 '10 at 14:31
2  
I note that this is now registered by "Stack Overflow Internet Services, Inc." – Rob Hyndman Jul 30 '10 at 13:18
6  
I downvoted this: it has no link to statistics for "outsiders", and as a statistician I don't get the "inside joke" feeling either. Cross-validation is a minor technique that is not that great either. – Aniko Aug 10 '10 at 19:09
4  
@Aniko: w.r.t. your point about the "inside joke", do you really think that everyone on stackoverflow.com knows what a stack overflow is? I am quite sure that they don't given that Jeff Atwood previously admitted that he had no idea what it was. I also don't entirely see how cross-validation is a minor technique; IMO, it's pretty fundamental, especially given the broad "data analysis" basis for the site. – Shane Aug 11 '10 at 23:41
1  
How about giving it an Italian flavour: CrossValidate.it ? – gd047 Aug 12 '10 at 12:00
8  
@Aniko. Every data miner would know about cross-validation, and I think most statisticians would too, although perhaps not under that name. After all, it is closely related to the jack-knife, and to leverage in regression, and to studentized residuals. And it is widely used in nonparametric regression for bandwidth selection. I disagree that it is a minor technique. – Rob Hyndman Aug 14 '10 at 3:27
1  
So my 2 cents is that it isn't a name that jumps out to me as being related to statistics. I've heard of all of the other things Rob lists above, but never had I heard them referred to as cross-validation until I came here. The litmis test for a name should be if you tell someone what the name is are they going to have a sense of what the site is about. Things may be different for others, but in my area of study I doubt they'll instantly know what the site "CrossValidated.com" is for or about. – drknexus Aug 17 '10 at 7:04
1  
The wikipedia disambiguation page for "cross validation" (en.wikipedia.org/wiki/Cross-validation) lists three meanings. Only one has anything to do with statistics, and that is a fairly narrow part of statistics. We can find a more general name. – Harvey Motulsky Aug 17 '10 at 23:38
2  
@(all that think one should be familiar with name to find/use the site) -- no, new users will find it by its contents, not name. As Shane mentioned stack overflow is a very specific error and is not commonly known, and still it makes no problem for SO. Also I don't think that median Windows user will now what superuser means. – mbq♦ Aug 18 '10 at 6:42
3  
drknexus - how is "StatsOverflow.com" any more memorable than "Crossvalidated.com"? "Overflow" only has meaning for people who know about StackOverflow, who form a very small group - and a group that's likely to be able to find a statistics site in the StackOverflow family without needing to know what the name is. Conversely, an advantage of CrossValidated is that it's clear about what kind of statistics are being talked about - it's clearly not sports/government/etc. statistics, it's analysis. – Matt Parker Aug 18 '10 at 14:45
1  
I would hope this site would welcome scientists who need help with data analysis. To facilitate this, we shouldn't use jargon in the site name. What would happen if asked a bunch of scientists to identify the field in which "cross-validate" is used, and offered five choices: a)statistics, b)psychotherapy, c)anthropology, d)accounting, or e)comparative literature. I suspect you'd end up with about 20% guessing each. – Harvey Motulsky Aug 18 '10 at 16:47
3  
@Harvey: (1) I think that you're wrong about that survey; this concept is primarily associated with statistics/machine learning. Again: just try a google search: all the top research relate to the right thing. (2) IMO, having jargon is a good thing, as per the blog post in the original question. (3) If you have a better idea, please propose it; I would be happy to see a better name myself (I like CV in part because of its meta interpretation). I see StatsOverflow as problematic both because it's "Stats" (hence not including other areas) and Overflow (which has nothing to do with us). – Shane Aug 18 '10 at 17:12
show 13 more comments
feedback
up vote 10 down vote

StatsOverflow.com

link
4  
Familiar, but wouldn't it be better to have a name that's unique to our endeavour? StackOverflow is a specific computer-science related concept. – Shane Jul 20 '10 at 13:13
2  
@Shane That is indeed a good point. I just liked the familiarity the name generates with stackoverflow, the same way mathoverflow.net does. – Tony Breyal Jul 20 '10 at 18:21
1  
@Tony That's why I think it should be something fitting into a S*[O*] scheme but still making sense. – mbq♦ Jul 23 '10 at 14:31
11  
It doesn't mean anything, unlike "StackOverflow". It will be harder to sell to people who aren't familiar with the SO site. – Rob Hyndman Jul 26 '10 at 9:29
7  
I like it. The "overflow" connotation has not hurt mathoverflow at all, and it is becoming a meme...i.e. people will understand what type of site it is immediately. – PeterR Jul 27 '10 at 19:30
5  
I just looked at some of the other SE sites, and I didn't see another one going for blahoverflow.com. I agree with @Rob that this has a specific meaning for stackoverflow, and the SO team even chose different names for superuser and serverfault that were more relevant. – Shane Aug 10 '10 at 20:48
1  
Although closely related, it attempts to borrow too much from it's successful progenitor. Although, arguably, many questions will be asked and answered that reflect the use of software for statistical analysis. – Brandon Bertelsen Aug 12 '10 at 3:54
1  
@Brandon Stack overflow is rather not the most hurting problem in statistical software... – mbq♦ Aug 16 '10 at 15:16
4  
Regarding the blog post linked in the original question, note #2: "Look for jargon that has meaning to the group of people you want to attract. Insider jargon is the duck call of insiders. Look at me! If I can say “contributory negligence,” I must be a real lawyer!" This speaks to the reason that I think this is a poor name selection. – Shane Aug 16 '10 at 17:31
5  
There's a Machine Learning proposal on Area 51 even though ML questions are part of the mission statement here. I think the "Statistical Analysis" name keeps some people away. We should go for a name that has meaning for both ML and Statistics folks to encourage more cross fertilization between the two areas. CrossValidated seems pretty good to me for that reason. – ars Aug 17 '10 at 5:48
feedback
up vote 8 down vote

MaximumLikelihood.com

link
feedback
up vote 6 down vote

StatsExchange.com

link
show 1 more comment
feedback
up vote 5 down vote

DegreesofFreedom

link
2  
I like this because it covers the data mining as well as the more traditional statistical side of the site. It also doesn't emphasise one area (e.g., Bayesian methods or hypothesis testing). – Rob Hyndman Jul 23 '10 at 0:11
1  
That it is not FreedomDegrees. – mbq♦ Jul 26 '10 at 5:26
show 1 more comment
feedback
up vote 5 down vote

ModelConvergence.com

link
feedback
up vote 5 down vote

OptimumModel.com

link
show 1 more comment
feedback
up vote 4 down vote

www.randomvariate.com

link
show 2 more comments
feedback
up vote 4 down vote

Overfitting.com

link
2  
Or maybe ModelOverfit.com to stick to two-word standard? – mbq♦ Jul 21 '10 at 10:03
2  
read as Mode-Lover-Fit also :-P – gd047 Aug 12 '10 at 11:19
show 1 more comment
feedback
up vote 4 down vote

ExplanatoryVariable.com

link
feedback
up vote 4 down vote

StatsFaq.com

I own this domain and would be happy to donate it. Nice and succinct, isn't it?

I also own the domains StatFaq.com, StatFaqs.com and StatsFaqs.com as forwarding addresses, so anyone who omits or adds an "s" would wind up at the right site.

link
show 2 more comments
feedback
up vote 3 down vote

AllYourBayes.com

If we don't want to appear partisan:

AllYourBayesAndFreqs.com

link
show 1 more comment
feedback
up vote 3 down vote

ModelOverfit.com

(available) Based on this discussion

link
feedback
up vote 3 down vote

EnsembleMethods.com

(available) This is a play on concept of ensemble learning:

In statistics and machine learning, ensemble methods use multiple models to obtain better predictive performance than could be obtained from any of the constituent models.

link
feedback
up vote 2 down vote

I listed a few here in the past, but my favorite is "Confidence Intervals"; several of these are available as domain names.

link
feedback
up vote 2 down vote

StatisticalSignificance.com

link
4  
I think we should try to avoid names that lean towards specific branches of statistics. For example, StatisticalSignificance and NullHypothesis have a frequentist slant to them. A bad choice of name may put many people off the site. – csgillespie Jul 20 '10 at 13:35
2  
We should also try to balance the data analysts/machine learning types who don't have any direct input in statistical debates. It's a tough balancing act to please everyone! – Shane Jul 20 '10 at 13:51
show 1 more comment
feedback
up vote 2 down vote

alternativehypothesis.com

link
show 1 more comment
feedback
up vote 2 down vote

ExploratoryAnalysis.com

link
feedback
up vote 2 down vote

RandomPatterns.com

link
feedback
up vote 1 down vote

Statosphere.com

link
show 1 more comment
feedback
up vote 0 down vote

PirateBayes.com

link
4  
Arrrrrrrrrrr... – Shane Jul 21 '10 at 13:05
1  
ThePirateBayes.com would be even better – Vebjorn Ljosa Jul 21 '10 at 14:00
1  
Amusing, but not a reasonable candidate given that it connotes a narrow focus on bayesian statistics alone. – Mike Lawrence Aug 7 '10 at 18:34
show 3 more comments
feedback
up vote -1 down vote

How about www.normallydistributed.com ?

link
2  
I don't love this because ideally our questions/answers will be fat tails (in terms of their degree of excellence). – Shane Jul 20 '10 at 13:12
1  
When I saw this, the smart@ss in me really wanted to submit www.abnormallydistributed.com but the serious side of me slapped the smart@ss side into submission. With that said and off of my chest, I agree with Shane. – A Lion Jul 20 '10 at 15:43
feedback
up vote -1 down vote

I thought of SignificanceTested. Domain is free, and it is in SE style.

link
show 1 more comment
feedback
up vote -1 down vote

MeanSquareErrors.com

link
1  
Regretfully, that's unavailable... – Shane Jul 20 '10 at 13:13
show 2 more comments
feedback
up vote -1 down vote

NullHypotheses.com

link
show 3 more comments
feedback
up vote -1 down vote

This one is available:

UnbiasedEstimates.com

link
feedback
up vote -1 down vote

BootstrapConfidence.com

link
feedback
up vote -1 down vote

infinitevariance.com

(Taken, but apparently for a personal site, so maybe the owner can be convinced to give it up.)

link
feedback
up vote -1 down vote

Totally Stats!

TotallyStats.com

Unambiguous. Easy to spell. Available.

link
2  
I just think that this is TotallyUncreative and TotallyBoring... :) Otherwise, it's awesome. – Shane Jul 27 '10 at 20:04
show 1 more comment
feedback
1 2

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged